A critically important part of understanding the disease process in T1D is the identification of proteins that stimulate diabetogenic T cells. The overall hypothesis underlying this project is that any autoantigen that contributes to pathogenesis is potentially both a therapeutic and diagnostic target. Thus one benefit to be realized from defining antigenic peptides for T cells is the provision of reagents that might usefully be applied to the induction of tolerance. Our objective is to take advantage of recent advances in methodology for biochemical isolation and identification of proteins (termed proteomics) to define autoantigens for pathogenic T cells that spontaneously arise in T1D in a well-established mouse model, and which might also be involved in the human disease. Specifically, we propose in this pilot application to identify the antigens for a panel of 6-8 islet-reactive, diabetogenic CD4+ T cell clones. Our aims are: (1) to identify candidate antigens for autoreactive T cells through a comparative analysis of total cellular, and selected organellar, proteomes obtained from stimulatory and non-stimulatory insulinomas; (2) to derive robust bioassays capable of the quantitative measurement of solubilized antigens in complex mixtures and use them to identify candidate antigens from pancreatic b-cells following preparative chromatographic and/or electrophoretic separations; and (3) to clone and express full-length or fragmented forms of the candidate proteins identified in aims 1 and/or 2 for confirmation of their specific immunogenicity.